Programming with RSK Sockets

Programming with the RSK socket library involves the following steps:

1. In each of your PL/X compilation units that will use the RSK socket library, you must include the RSK socket library language binding macro. To do so, put the following statement into each compilation unit:

   %include syslib(plxsock);

   PLXSOCK COPY is in DMSRP MACLIB, which is part of the z/VM® PL/X Restricted Source Feature, which you can order as a feature of z/VM.
2. At run-time, your first step must be to initialize the RSK socket library. This prepares the library to receive socket calls. To initialize the library, you must either call PS_libinit() yourself or arrange for the RSK to call it. See PS_libinit for more information.
3. To perform socket operations, you must create a socket set. ¹ We call each RSK socket set an application and hence the entry point you use for this is PS_applinit().

   You supply PS_applinit() with the name (VM user ID) of the TCP/IP stack machine, a unique name for your new set of sockets, and the number of sockets you want in the set.

   PS_applinit() establishes the IUCV connection to the TCP/IP stack machine and prepares the socket set for your use.

4. You perform operations on the sockets in your set. You use the RSK socket library entry points to do so. For example, to allocate a new socket, you call PS_socket(), or to write data to a socket, you call PS_write().
5. When you are done with your set of sockets, you dispose of it by calling PS_applterm(), identifying the socket set by the unique name you chose for it at its creation.
6. Prior to your server ending, either you should call PS_libterm() or you should arrange for the RSK to call it. See PS_libterm for more information.